#fractional-bit precision

PolyQ: Codesigning End-to-End Quantization Framework for Scalable Edge CPU LLM Inference

Hyunwoo Oh, Suyeon Jang, Hanning Chen +4

PolyQ is a co-designed compiler and quantization framework that assigns per‑channel bit‑widths to LLM activations on CPUs, enabling fine‑grained fractional‑bit precision while keep…